Documentation >
Glossary R
Glossary R
The financial world is full of jargon -
i.e. strange words no-one understands. Here we
try to explain some of the many technical terms.
A | B | C | D | E | F | G | H | I | J | K | L | M | N | O | P | Q | R | S | T | U | V | W | X | Y | Z
Ramping
Logged-on to your favourite bulletin board, you
see a message from someone whose views you have
grown to respect; 'Get in on stock ABC, its going
to go ballistic ...' - the temptation is just too
great and you buy some stock, as do many others;
the stock does indeed shoot up, then it crashes
back down again very sharply, and in the end you
have lost a tidy sum.
What happened was that your friend, the poster,
had a load of dud stock he wanted to get rid of
at a profit - when the price rose in response to
his message, he quickly sold the shares he owned.
Posting rumours on the internet to influence
share prices was an early tactic (- think of all
the possible people one can contact via the
web!), but is becoming rarer. Most of the
bulletin boards are moderated and will be aware
of these kind of stratagems. Ramping
stock/posting false rumours can also get you into
trouble with the law; you could face criminal
charges and even if you avoid this, there is the
possibility of facing a large civil damages suit
from your victim, especially if you bad-mouthed
one of the big guys.
If you still think this is a 'great idea', and
want to try it, remember that there is no such
thing as anonymity on the web - at least for the
ordinary surfer; everything you do is logged
somewhere, and you can be traced.
The property of having no
pattern, no possible simplification, no
underlying order among a set of objects. Random
objects have no simpler 'explanation' than
themselves.
In relation to timeseries, while it is possible
to take any data sample whatsoever and fit a
curve to it, if the underlying process which
generated the data is itself random, then the
fitted curve will have no predictive
value whatsoever. Let us emphasize this -
random processes cannot be
predicted - this is their defining
feature; they can be categorised by their
probability density function, but firm prediction
is not possible. A practical test for the
existence of randomness is compressibility; if an
object is random, you cannot compress it.
When dealing with processes which have a large
random component we switch to a probabilistic
mode of thinking i.e. instead of calculating what
something will be, we calculate what the
probability (chance) of an event is.
Philosophically, this is uncomfortable for many
people - 'things happen for a reason/everything
is caused by something else' - the realization
that randomness exists in nature at a fundamental
level has been crucial to the development of
science.
Lots of randomness around when we look at
stock prices.
Random Walk
Imagine a drunk man trying to get home; he does
not know where he is, has no memory of where he
has been, and does not even have a destination in
mind - he simply takes a step forward and then a
lurch to either side. Who knows where he will end
up! No one can predict this - even he has no
idea, but you can calculate the probability of
where he will end up after a number of steps. The
technique for doing this is called monte carlo
simulation; what we do is to imagine his walk
home many times, generating his sideways lurches
according to the relative frequency of their
occurrence; after many runs we can see how many
times he ended up in a certain place - this is
our probability; and this is also how we
should bet if for some reason we were
watching this process with a group of others and
gambling against them.
The random meanderings of a drunkard and the
movements of share prices should seem pretty
similar to you by now having studied our charts;
we hope you understand our analogy, and see that
it makes sense. Of course, it is only a first
step in our analysis - not the end of it, and
leaves many questions unanswered; this is where
we have to work much harder and do some clever
things to deal with external events and the
possibility of time correlations. Going back to
our drunk man, suppose we knew he was entering
territory where he may be waylaid by thugs,
attacked by an ill-tempered local hound,
encounter a kebab shop and feel hungry, or does
in fact retain some memory of his actions - we
still need to calculate his probable destination,
and we can - but the problem is now much more
difficult, and will be changed from that of our
initial model.
In relation to the stock market, the situations
facing the drunk are analogous to, e.g. interest
rate cuts, corporate scandals, natural disasters,
war and investor-driven speculative instability.
Ratio Call Spread
An options strategy; you buy 2 higher strike
price calls and sell 1 lower strike price call.
Ratio Put Spread
An options strategy; you buy 1 higher strike
price put and sell two lower strike price puts.
Real-time Quotes
Not delayed.
Recurrent
A technical term concerning neural network
architectures; recurrent networks have feedback
incorporated into their training algorithms -
sometimes this improves performance, sometimes it
is simply extra complication for little
advantage.
Reform, Reforms
Destroy.
Destructive
measures, ostensibly reasonable, benign,
positive. [Orwellian]
Religion
A business that does not pay tax.
If you want to be really, really rich, your best
chance is to form your own religion; most human
beings have an in-built need to believe in
something, so why not take advantage of this.
Remuneration
Committee
A bunch of company directors who get together in
the pursuit of fairness, accountability and
external objectivity, to decide what the salary
of some other company director should be. Someone
shouts a telephone number, then another shouts -
'sounds like the market rate to me' - and the
issue is settled, albeit for minor trifles like -
college fund for the grandchildren, free medical
and dental for all family members for life, use
of company facilities, etc etc.
Top company guys get paid top wages, some of them
may even deserve it. The remuneration committee
is there to make sure that these very substantial
rewards are 'fair'. But in practice, these
committees hardly ever veto a proposed package,
simply because it would set a precedent which
could come back to damage its members in the
future.
Reporting Dates
The dates on which companies issue their reports;
can be yearly, half-yearly or quarterly. There
can be a lot of volatility around these times.
Reserve Currency
A default, backup, baseline currency - something
which can always be used, or what some commodity
is traded in. Having your own currency as a
reserve currency gives you a lot of power over
other countries and their economies.
Respectable Scams
Burglars broke into your house while you were
away; wearing masks and in broad daylight, they
backed a truck up to your front door and carried
off all your stuff ... !
- this is one way of stealing from people.
Another way is to wear a sharp suit, grin a
trusting smile and get them to give you their
money, once a month by direct debit, by selling
them a with-profits life assurance
policy, or an endowment
mortgage, or a private
pension, or a high income
bond, or a split capital
trust.
The smarter criminals use this second method as
there is no chance of doing jailtime there. The
first method is at least more honest and you will
be covered by your insurance policy; once
recovered from the shock, you will get your money
back (- or perhaps even a little extra, you
naughty person!); in the second case you will
never see your money again.
Return on Equity
Return on equity is calculated by taking a year's
worth of earnings and dividing them by the
average shareholder's equity for that year. One
of the quickest ways to gauge whether a company
is an asset creator or a cash consumer is to look
at the return on equity that it generates.
Return on equity (ROE), return on assets (ROA)
and return on investment (ROI) are all common
ratios used to measure, in broad terms, just how
good a company is at taking money and
generating more money from it; you can get
into nitpicking arguments about which one is best
at measuring the strength of a company, but I
wouldn't bother with them.
Retail Investor
The small private investor - you, me, all the
little guys. As opposed to what is presumably,
the wholesale investor - Pension Funds, Life
Insurance Companies and Investment Banks.
Retained Earnings
Retained earnings measures the amount of capital
a company has generated and is best used to
determine what sorts of returns on capital a
company has produced.
Return on Assets
The last 12 months net income divided by the
total assets from the most recent quarter. This
is a measure of effectiveness in using the assets
at hand in generating earnings.
Rocket Science
Any moderately advanced mathematical techniques
are routinely referred to as 'rocket science' by
the mass media; often applied to the activities
of quantitative analysts in developing derivative
pricing models.
StockWave is loaded with rocket science.
Rumour
What you get on the bulletin boards and the
newsgroups.
Be very wary of using these information sources -
the classic situation is that of
ramping or pump and dump, see
scams.
A | B | C | D | E | F | G | H | I | J | K | L | M | N | O | P | Q | R | S | T | U | V | W | X | Y | Z
Tutorials
FAQ
Copyright © 2006,2007 StockWave Software Ltd. All
Rights Reserved.
|